P284 - ASSESSMENT OF GLIM CRITERIA FOR MALNUTRITION AND PROGNOSTIC SIGNIFICANCE IN HEPATOCELLULAR CARCINOMA PATIENTS RECEIVING COMBINATION IMMUNOTHERAPY

P284

ASSESSMENT OF GLIM CRITERIA FOR MALNUTRITION AND PROGNOSTIC SIGNIFICANCE IN HEPATOCELLULAR CARCINOMA PATIENTS RECEIVING COMBINATION IMMUNOTHERAPY

A. Kitaoka^1,*, K. Oura², A. Morishita², H. Kobara²

¹Division of Clinical Nutrition and Food Services, Kagawa University Hospital, Faculty of Medicine, ²Department of Gastroenterology and Neurology, Faculty of Medicine, Kagawa University, Kagawa, Japan

Rationale: Malnutrition and muscle wasting are linked to poor prognosis in advanced hepatocellular carcinoma (HCC). Conventional assessments may be affected by liver function and comorbidities.The GLIM criteria offer a standardized malnutrition diagnosis, but data in HCC patients on systemic therapy are limited. This study compared the clinical utility of GLIM and SGA in HCC patients receiving combination immunotherapy.

Methods: We retrospectively analyzed 108 patients with unresectable HCC treated with atezolizumab plus bevacizumab or durvalumab plus tremelimumab (2020–2024). Malnutrition was diagnosed by GLIM and SGA. Using SGA as the reference standard, the diagnostic accuracy of GLIM was evaluated with sensitivity, specificity, AUC, and κ coefficients. We also assessed adverse events, treatment response, and survival.

Results: Mean age was 74.4 ± 9.9 years; 73.4% were male. Of 108 patients, 42 (38.9%) were diagnosed with malnutrition according to GLIM, and 48 (44.4%) according to SGA.GLIM showed sensitivity of 83.3%, specificity of 80.3%, and AUC of 0.818.The odds ratio compared to SGA was 20.38 (95% CI: 7.4–56.1, p < 0.01), and the κ coefficient was 0.620.GLIM-defined malnutrition was associated with more adverse events and poorer treatment response. The median PFS was 263 days in the malnourished group and 716 days in the well-nourished group, while OS was 345 and 716 days, respectively (p < 0.05). GLIM provided better prognostic discrimination than SGA and was independently associated with OS in multivariate analysis.

Conclusion: GLIM criteria showed good agreement with SGA and were effective for nutritional assessment of HCC patients receiving combined immunotherapy. GLIM also showed good prognostic value and may help identify at-risk patients in clinical practice.

Disclosure of Interest: None declared